Finance Agent (Vals AI) leaderboard
- Gemini 4 Argon leads
- 65.4%
- models measured
- 75
- lead over #2
- +4.0 pts
- median
- 51.0%
Leader on Finance Agent (Vals AI) · read October 21
On October 21, 2026, 75 of our catalog's models had a Finance Agent (Vals AI) result, and Google Gemini 4 Argon led with 65.4%.
Research tasks a financial analyst does, worked by an agent with tools. Share answered correctly, as Vals AI grades it.
Role in the index
We collect this benchmark and show it as supporting evidence. It is not part of the Vibe Coding Index, because it measures something other than coding and the index is about coding. Read the methodology
| # | Model | Result | Cost per test | Duration, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Gemini 4 ArgonGoogle | 65.4% | $4.38 | 16m 21s | 80.5 | $10.00 | |
| 2 | Gemini 3.8 FlashGoogle | 61.4% | $2.00 | 3m 21s | 70.5 | $4.125 | |
| 3 | Muse Spark OpenMeta | 61.3% | $0.77 | 24m 13s | 62.1 | $4.25 | |
| 4 | Muse Spark 1.2Meta | 60.6% | $0.77 | 5m 04s | 65.1 | $4.25 | |
| 5 | Gemini 3.7 FlashGoogle | 59.0% | $1.48 | 3m 01s | 69.9 | $3.75 | |
| 6 | Muse Spark 1.3Meta | 58.9% | $0.74 | 5m 42s | 79.7 | $4.25 | |
| 7 | Claude Fable 5.1Anthropic | 58.9% | $8.35 | 18m 14s | 89.1 | $50.00 | |
| 8 | Claude Opus 5Anthropic | 58.6% | $5.12 | 9m 58s | 81.8 | $25.00 | |
| 9 | Claude Opus 5.5Anthropic | 58.6% | $9.22 | 33m 08s | 92.2 | $20.00 | |
| 10 | Claude Sonnet 5.5Anthropic | 58.1% | $6.55 | 31m 43s | 88.9 | $10.00 | |
| 11 | Gemini 3.5 FlashGoogle | 57.9% | $2.51 | 5m 22s | 42.8 | $9.00 | |
| 12 | GLM 5.3 FlashZ.ai | 57.9% | $0.05 | 14m 51s | 71.0 | $0.50 | |
| 13 | MiMo-V2.6-ProXiaomi | 57.3% | $0.20 | 10m 48s | 72.6 | $0.87 | |
| 14 | Muse Spark 1.1Meta | 57.2% | $0.72 | 6m 55s | 48.9 | $4.25 | |
| 15 | DeepSeek V4.1 Flash 1017DeepSeek | 57.0% | $0.46 | 14m 39s | 83.3 | $0.60 | |
| 16 | Muse WonderMeta | 56.6% | $2.67 | 6m 09s | 64.5 | $20.00 | |
| 17 | Claude Fable 5Anthropic | 56.3% | $8.06 | 10m 11s | 82.1 | $50.00 | |
| 18 | Gemini 3.6 FlashGoogle | 56.3% | $1.40 | 3m 45s | 48.3 | $3.75 | |
| 19 | MiMo-V2.6-FlashXiaomi | 56.3% | $0.07 | 7m 02s | 61.9 | $0.28 | |
| 20 | GLM 5.3Z.ai | 55.8% | $1.07 | 15m 50s | 73.1 | $1.40 | |
| 21 | Hy4 previewTencent | 55.1% | $0.59 | 23m 45s | 61.6 | $2.501 | |
| 22 | GPT-5.6 LunaOpenAI | 55.0% | $0.28 | 12m 52s | 61.5 | $1.20 | |
| 23 | Mistral Large 4Mistral AI | 54.7% | $1.20 | 26m 02s | 55.3 | $4.18 | |
| 24 | GPT-5.6 TerraOpenAI | 54.4% | $3.63 | 24m 10s | 69.0 | $12.00 | |
| 25 | Claude Haiku 5.5Anthropic | 54.1% | $1.64 | 30m 49s | 75.9 | $0.50 | |
| 26 | Claude Opus 4.8Anthropic | 53.9% | $4.22 | 9m 07s | 66.0 | $25.00 | |
| 27 | Claude Sonnet 5Anthropic | 53.9% | $0.75 | 13m 12s | 63.5 | $10.00 | |
| 28 | GPT-5.6 SolOpenAI | 53.8% | $1.25 | 19m 25s | 76.7 | $20.00 | |
| 29 | Grok 4.6xAI | 53.7% | $1.66 | 16m 41s | 76.4 | $6.00 | |
| 30 | GPT-6 AstraOpenAI | 53.5% | $6.82 | 11m 33s | 88.7 | $50.00 | |
| 31 | DeepSeek V4.1 FlashDeepSeek | 53.5% | $0.21 | 6m 04s | 77.2 | $0.525 | |
| 32 | Kimi K3Moonshot AI | 53.1% | $1.91 | 4m 46s | 61.8 | $11.936 | |
| 33 | Grok 4.7xAI | 52.3% | $2.72 | 15m 33s | 75.0 | $6.00 | |
| 34 | GPT-6.1 SolOpenAI | 52.0% | $1.62 | 15m 25s | 88.7 | $10.00 | |
| 35 | Ember-1Fireworks | 51.8% | $1.95 | 13m 19s | Not benchmarked | $15.00 | |
| 36 | GPT-5.5OpenAI | 51.8% | $4.15 | 11m 02s | 67.8 | $30.00 | |
| 37 | Claude Opus 4.7Anthropic | 51.5% | $4.03 | 6m 00s | 51.1 | $25.00 | |
| 38 | Claude Sonnet 4.6Anthropic | 51.0% | $2.41 | 11m 36s | 57.8 | $15.00 | |
| 39 | Step 5 PreviewStepFun | 50.7% | $0.62 | 12m 28s | 71.4 | $2.70 | |
| 40 | Qwen3.8 Max (0902)Qwen | 50.6% | $1.24 | 19m 35s | 63.5 | $6.00 | |
| 41 | DeepSeek V4 Pro 0813DeepSeek | 50.4% | $0.88 | 16m 15s | 54.6 | $1.98 | |
| 42 | GPT-6 LunaOpenAI | 49.9% | $0.12 | 14m 58s | 71.4 | $0.50 | |
| 43 | GLM 5.2Z.ai | 49.7% | $0.71 | 7m 53s | 57.7 | $7.00 | |
| 44 | DeepSeek V4 Flash 0731DeepSeek | 49.5% | $0.28 | 8m 10s | Not benchmarked | $1.32 | |
| 45 | GPT-6 SolOpenAI | 49.0% | $2.12 | 9m 00s | 82.8 | $10.00 | |
| 46 | Qwen3.8 27BQwen | 48.5% | $0.75 | 10m 22s | 73.6 | $1.35 | |
| 47 | Grok 4.5xAI | 48.4% | $1.16 | 4m 49s | 75.5 | $6.00 | |
| 48 | MiniMax M3MiniMax | 48.3% | $0.32 | 8m 17s | 57.8 | $1.20 | |
| 49 | Qwen3.7 MaxQwen | 47.8% | $0.55 | 10m 35s | 45.1 | $4.425 | |
| 50 | Gemini 3.5 Flash LiteGoogle | 47.4% | $0.39 | 2m 23s | 33.6 | $2.50 | |
| 51 | T InklingThinking Machines | 46.6% | $1.00 | 18m 02s | 49.7 | $4.05 | |
| 52 | GPT-5.4 MiniOpenAI | 45.4% | $1.20 | 35m 54s | 28.2 | $4.50 | |
| 53 | Kimi K2.6Moonshot AI | 44.9% | $1.14 | 27m 10s | 44.2 | $3.41 | |
| 54 | GLM 5.1Z.ai | 44.8% | $0.62 | 15m 24s | 44.0 | $4.40 | |
| 55 | DeepSeek V4 Pro 0423DeepSeek | 44.1% | $0.88 | 25m 26s | 35.3 | $1.9105 | |
| 56 | Gemini 3.1 Pro PreviewGoogle | 43.0% | $1.59 | 3m 28s | 35.3 | $12.00 | |
| 57 | Gemini 3 Flash PreviewGoogle | 42.5% | $0.57 | 4m 07s | 75.8 | $3.00 | |
| 58 | MiMo-V2.5-ProXiaomi | 41.5% | $0.21 | 8m 17s | 40.0 | $0.609 | |
| 59 | T Inkling SmallThinking Machines | 41.3% | $0.23 | 16m 34s | 29.9 | $1.20 | |
| 60 | Qwen3.6 PlusQwen | 40.9% | $0.81 | 10m 12s | 37.4 | $1.95 | |
| 61 | GPT-5.4 NanoOpenAI | 38.2% | $0.16 | 5m 35s | Not benchmarked | $1.25 | |
| 62 | Qwen3.7 PlusQwen | 38.2% | $0.36 | 9m 02s | Not benchmarked | $1.28 | |
| 63 | Grok 4.3xAI | 37.7% | $0.85 | 10m 15s | 22.5 | $2.50 | |
| 64 | Nemotron 3 UltraNVIDIA | 37.7% | Not published | 4m 14s | Not benchmarked | $2.40 | |
| 65 | MiMo-V2.5Xiaomi | 36.7% | $0.09 | 5m 39s | 34.0 | $0.238 | |
| 66 | Kimi K2.5Moonshot AI | 35.8% | $0.29 | 12m 32s | 58.5 | $2.50 | |
| 67 | Mistral Medium 3.5Mistral AI | 32.1% | $1.91 | 12m 48s | 9.2 | $7.50 | |
| 68 | Claude Haiku 4.5Anthropic | 31.0% | $0.60 | 2m 57s | 49.8 | $5.00 | |
| 69 | Ling 3.0 FlashinclusionAI | 30.3% | $0.05 | 2m 37s | Not benchmarked | $0.063 | |
| 70 | Gemini 3.1 Flash Lite PreviewGoogle | 30.0% | $0.14 | 55 s | 8.1 | $1.50 | |
| 71 | Grok 4.20xAI | 28.5% | $0.79 | 5m 51s | Not benchmarked | $2.50 | |
| 72 | MiniMax M2.7MiniMax | 27.9% | $0.18 | 13m 24s | 28.4 | $1.20 | |
| 73 | Mercury 2.5Inception | 18.9% | $0.11 | 1m 32s | Not benchmarked | $0.15 | |
| 74 | Nemotron 3.5 LightningNVIDIA | 18.5% | $0.03 | 9m 00s | Not benchmarked | $0.20 | |
| 75 | Command ACohere | 9.0% | Not published | 9m 27s | Not benchmarked | $10.00 |
Only models in our catalog are listed, and each appears once with its best published run. 132 rows were read from the source on October 21, 2026, and rows that match no catalog model are left out. Cost per test is what Vals AI reports one test cost on the run behind the result, in US dollars. Every other column is a figure the board publishes for the same run.